Ball valve body rotation stopping method and system of two-way stratum isolation valve

Abstract

The invention provides a ball valve body rotation stopping method and system of a two-way stratum isolation valve, and belongs to the field of oil and gas wells, and the method comprises the steps that environment data of a ball valve body and sensors around the ball valve body are obtained, all the environment data are preprocessed, and an environment state data set is generated; according to the position finding requirement of the ball valve body, a reinforcement learning model is constructed, the environment state data set serves as input, and optimal position finding parameters of the ball valve body are output; current underground working condition characteristics of the environment state data set are extracted, self-adaptive adjustment is conducted on the optimal position finding parameters based on the current underground working condition characteristics, and the optimal rotation stopping torque is generated; constructing a sealing state prediction model according to the optimal rotation stopping torque and the sealing monitoring key variables of the environment state data set, evaluating the sealing quality and predicting the leakage trend, detecting abnormal conditions, performing intelligent fine adjustment, and generating optimal rotation stopping parameters; and generating an optimal position finding parameter corresponding to the adjustment target based on the optimal rotation stopping torque and the rotation stopping torque adjustment amount.

Description

Technical Field

[0001] The present invention belongs to the field of ball valve body rotation and stop, and particularly relates to a method and system for rotating and stopping the ball valve body of a two-way formation isolation valve. Background Art

[0002] In the cementing and isolation operations of oil and gas wells, the formation isolation valve is a key piece of equipment. Its main function is to achieve fluid isolation downhole, prevent fluid crossflow between different formations, and provide necessary isolation support to ensure the safety and effectiveness of downhole operations. One of the core components of the formation isolation valve is the ball valve body, and its rotation and stop actions determine the sealing effect and service life of the valve. Existing ball valve body rotation and stop technologies usually rely on mechanical positioning and fixed torque control methods, that is, downhole, through mechanical structures such as limit slots and claws, the ball valve body is docked with the sealing surface, and a fixed torque is applied to complete the rotation and stop. However, in actual applications, due to the complex downhole conditions, existing technologies have defects in many aspects.

[0003] First of all, there are large errors in the mechanical positioning method. Under the influence of environments such as high temperature, high pressure, and fluid erosion, the ball valve body may not be accurately docked with the sealing surface, resulting in a decrease in the sealing effect and even leakage. Since the traditional positioning mechanism usually uses preset mechanical limit structures, the changes in downhole temperature and pressure may cause these structures to deform, further reducing the positioning accuracy.

[0004] In addition, the rotation and stop action of the ball valve body depends on the fixed torque setting. However, due to the different requirements for the rotation and stop torque in different downhole environments, the fixed torque may cause damage to the sealing surface or insufficient sealing. In a high-pressure environment, excessive torque may cause plastic deformation of the sealing surface, affecting the long-term sealing ability; while in a low-pressure or high-temperature environment, insufficient torque may result in insufficient sealing, increasing the leakage risk. In addition, existing technologies usually lack an intelligent feedback mechanism for the sealing state. After the rotation and stop of the ball valve body are completed, it cannot be dynamically optimized according to the sealing situation. Once the sealing is poor, additional remedial measures such as repeated rotation and stop or additional isolation treatment are often required, which not only increases the construction complexity but also may affect the entire operation process.

[0005] Therefore, the current ball valve body rotation and stop methods still have great room for improvement in terms of positioning accuracy, rotation and stop torque control, and sealing state optimization, and a more intelligent technology is needed to improve the reliability and adaptability of rotation and stop. Summary of the Invention

[0006] The object of the present invention is to propose a method and system for rotating and stopping the ball valve body of a two-way formation isolation valve, which can train the rotation and stop strategy in real time through downhole sensing data, enabling the ball valve body to adjust the rotation angle according to the actual downhole conditions and improving the accuracy of sealing docking.

[0007] To achieve the above object, in a first aspect, the present invention provides a method for rotating and stopping the ball valve body of a two-way formation isolation valve, the method comprising:

[0008] Obtain the environmental data of the ball valve body and the sensors around it, and preprocess all the environmental data to generate an environmental state data set;

[0009] According to the positioning requirements of the ball valve body, construct a reinforcement learning model, take the environmental state data set as input, and output the optimal positioning parameters of the ball valve body; wherein, the optimal positioning parameters include: the optimal rotation angle and the optimal positioning torque;

[0010] Extract the current downhole working condition characteristics of the environmental state data set, and adaptively adjust the optimal positioning parameters based on the current downhole working condition characteristics to generate the optimal rotation and stop torque;

[0011] According to the optimal rotation and stop torque and the key variables for seal monitoring in the environmental state data set, construct a seal state prediction model, evaluate the seal quality and predict the leakage trend, detect abnormal conditions and perform intelligent fine-tuning to generate the optimal rotation and stop parameters;

[0012] Calculate the fine-tuning rotation and stop torque adjustment amount based on the adaptive adjustment target and the evaluation result of the seal quality, and generate the optimal positioning parameters corresponding to the adjustment target based on the optimal rotation and stop torque and the rotation and stop torque adjustment amount;

[0013] Wherein, the adaptive adjustment target is seal pressure adjustment.

[0014] Preferably, the environmental data includes downhole real-time pressure, the vibration amplitude of the ball valve body, the current torque of the ball valve body, and the shear force of the downhole fluid; the preprocessing includes:

[0015] Obtain the original environmental data;

[0016] Smooth the continuous data points of the original environmental data, wherein the window size is dynamically adjusted through historical data;

[0017] Detect the outliers in the data after smoothing processing, and interpolate and compensate the data points beyond the range to obtain the corrected data;

[0018] Extract the key features of the corrected data and perform normalization processing; wherein, the key features include the maximum value, minimum value, mean value, and variance within the current data window;

[0019] Obtain the preprocessed data and construct it into an environmental state data set.

[0020] Preferably, the maximum value, minimum value, mean value, and variance within the current data window include:

[0021] The maximum value, minimum value, average value, and variance of the downhole real-time pressure; the maximum value, minimum value, average value, and variance of the vibration amplitude of the ball valve body, and the current torque of the ball valve body; the maximum value, minimum value, average value, and variance of the current torque of the ball valve body.

[0022] Preferably, constructing a reinforcement learning model according to the positioning requirements of the ball valve body specifically includes:

[0023] Obtaining an environmental state dataset, constructing a reinforcement learning model, and adjusting the strategy when using the environmental state dataset to enable the ball valve body to accurately dock with the sealing surface under changing downhole conditions; setting the rotation angle and positioning torque of the ball valve body as control variables, and initializing the reinforcement learning intelligent positioning strategy of the reinforcement learning model;

[0024] At the current time step, calculate the current environmental state by the reinforcement learning model, select the optimal positioning action based on the current strategy, and after applying the positioning action, collect the docking deviation and sealing pressure of the ball valve sealing surface to judge the positioning effect;

[0025] Calculate the reward value based on the state data after positioning, and introduce a positioning stability regularization term to optimize the positioning strategy;

[0026] Among them, the positioning strategy satisfies the following constraints:

[0027] A. For high-pressure and high-flow conditions, it is necessary to enhance the positioning torque compensation to prevent positioning failure caused by fluid impact;

[0028] B. For low-sealing conditions downhole, a step-by-step positioning method needs to be adopted to gradually adjust the rotation angle until the sealing pressure reaches the target value;

[0029] Among them, when the docking deviation of the ball valve sealing surface is less than or equal to the preset tolerance and the sealing pressure is greater than or equal to the sealing pressure threshold value, the positioning is successful, and the optimal positioning parameters are output: the optimal rotation angle and the optimal positioning torque.

[0030] Preferably, collecting the docking deviation and sealing pressure of the ball valve sealing surface to judge the positioning effect specifically includes:

[0031] If the docking deviation of the ball valve sealing surface is equal to the difference between the current rotation angle and the target rotation angle, it is directly used to evaluate the angle error of the current positioning;

[0032] If the docking deviation of the ball valve sealing surface is greater than the preset tolerance, adjust the rotation angle and update the strategy;

[0033] If the sealing and fitting pressure is less than the sealing and fitting pressure threshold value, it indicates that the positioning torque may be insufficient, and the positioning torque needs to be increased for compensation.

[0034] Preferably, adaptively adjusting the optimal positioning parameters based on the current downhole working condition characteristics to generate the optimal rotation stop torque specifically includes:

[0035] Set the initial torque for the rotation stop of the ball valve , and dynamically correct it based on the environmental state data to ensure that the initial torque adapts to the current downhole working conditions, and the calculation is as follows:

[0036]

[0037] Wherein, is the standard downhole pressure, is the reference fluid shear force; and are the pressure and shear force compensation coefficients respectively; Apply the torque to rotate and stop the ball valve body, and real-time monitor the seal state parameters:

[0038] Sealing pressure , which is used to judge the seal state.

[0039] Leak detection signal , which is used to judge whether there is seal failure;

[0040] Monitor the seal state parameters. If the sealing pressure is lower than the set threshold value or the leak detection signal , it indicates that there is a leak, then adjust the torque:

[0041]

[0042] Wherein, is the torque adjustment amount, is the sealing pressure compensation coefficient; is the leak correction coefficient;

[0043] Combine the initial torque and the torque adjustment amount to calculate the final optimal rotation stop torque. When the sealing pressure reaches the set value and there is no leak signal, the rotation stop is successful, and the optimal rotation stop torque is output.

[0044] Preferably, the seal state prediction model is predicted based on the leakage risk regularization term to ensure that the leakage risk can be pre-warned in advance near the seal state critical point, and the seal quality score is output;

[0045] If the seal quality score is greater than the set value, trigger an intelligent warning and analyze the change trend of the seal parameters:

[0046] Calculate the change amount of the sealing pressure and judge whether the sealing performance is gradually decreasing;

[0047] Judging the seal failure type based on long-term trends and short-term changes:

[0048] If the change in seal pressure gradually decreases and the leak detection signal increases, material fatigue may occur on the seal surface, and the rotation stop torque needs to be increased;

[0049] If the change in seal pressure suddenly decreases and the leak detection signal is 1, microcracks may occur on the seal surface, and intelligent fine-tuning needs to be performed to recalculate the optimal rotation stop parameters.

[0050] Preferably, the intelligent fine-tuning and recalculation of the optimal rotation stop parameters specifically include:

[0051] Setting the adaptive adjustment target: If the seal fitting pressure is less than the seal fitting pressure threshold, the rotation stop torque needs to be increased; if the seal fitting pressure is greater than the seal fitting pressure threshold and the leak detection signal is too large, there may be seal surface overload or uneven mechanical stress, and the rotation stop torque needs to be reduced and the rotation angle needs to be adjusted;

[0052] Calculating the adjustment amount of the fine-tuning rotation stop torque :

[0053]

[0054] where, is the seal pressure correction coefficient; is the vibration influence correction coefficient; is the seal fitting pressure threshold; is the seal fitting pressure; is the leak detection signal; Calculating the adjustment amount of the rotation angle fine-tuning :

[0055]

[0056] where, is the leak signal trigger adjustment coefficient; is the leak trend correction coefficient; is the leak trend prediction value, calculated based on time series analysis;

[0057] Applying the adjusted rotation stop torque and the adjusted rotation angle, re-monitoring the seal state, and judging whether the target is reached, then iteratively executing the fine-tuning process until the seal is stable.

[0058] In a second aspect, the present invention also provides a rotation stop system for the ball valve body of a two-way formation isolation valve, and the system includes:

[0059] A ball valve body data acquisition unit, configured to acquire the environmental data of the ball valve body and the sensors around it, and preprocess all the environmental data to generate an environmental state data set;

[0060] A positioning parameter analysis unit, configured to construct a reinforcement learning model according to the positioning requirements of the ball valve body, take the environmental state data set as an input, and output the optimal positioning parameters of the ball valve body; wherein, the optimal positioning parameters include: an optimal rotation angle and an optimal positioning torque;

[0061] A rotation stop torque analysis unit, configured to extract the current downhole working condition characteristics of the environmental state data set, adaptively adjust the optimal positioning parameters based on the current downhole working condition characteristics, and generate an optimal rotation stop torque;

[0062] An intelligent fine-tuning unit, configured to construct a seal state prediction model according to the optimal rotation stop torque and the key variables for seal monitoring in the environmental state data set, evaluate the seal quality and predict the leakage trend, detect abnormal conditions and perform intelligent fine-tuning to generate optimal rotation stop parameters;

[0063] A ball valve body rotation stop unit, configured to calculate a fine-tuning rotation stop torque adjustment amount based on the adaptive adjustment target and the evaluation result of the seal quality, and generate the optimal positioning parameters corresponding to the adjustment target based on the optimal rotation stop torque and the rotation stop torque adjustment amount;

[0064] Wherein, the adaptive adjustment target is seal pressure adjustment

[0065] The beneficial technical effects of the present invention are at least as follows:

[0066] Aiming at the deficiencies of the prior art, the present invention provides a ball valve body rotation stop method based on intelligent learning and adaptive control, which can improve the positioning accuracy, realize the adaptive adjustment of the rotation stop torque, and optimize the seal state. The present invention adopts an intelligent learning method to train the rotation stop strategy in real time through downhole sensing data, enabling the ball valve body to adjust the rotation angle according to the actual downhole working conditions and improving the accuracy of seal docking. Aiming at the problem of large errors in traditional mechanical positioning methods, the present invention combines various downhole sensing data, such as pressure, vibration, shear force, etc., to establish an intelligent positioning system, enabling the ball valve body to automatically fine-tune to the best docking position before rotation stop to ensure seal reliability. During the rotation stop process, the present invention adopts an adaptive control method to enable the rotation stop torque to be dynamically adjusted according to changes in the downhole environment, avoiding problems such as seal surface damage or insufficient sealing that may be caused by the fixed torque method. The present invention also provides an intelligent feedback mechanism. After the ball valve body rotation stop is completed, the system will continuously monitor the seal state. If poor sealing is found, the rotation stop angle or torque can be automatically adjusted to optimize the seal effect and improve the long-term seal stability. Through the above technical innovations, the present invention can effectively improve the positioning accuracy and seal reliability of the ball valve body, reduce human intervention, increase the service life and construction efficiency of the formation isolation valve, and is applicable to various complex downhole operation environments. Description of the Drawings

[0067] The present invention will be further described with reference to the accompanying drawings. However, the embodiments in the drawings do not constitute any limitation to the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the following drawings without creative efforts.

[0068] Figure 1 It is a flow chart of the ball valve body rotation stop method for a two-way formation isolation valve of the present invention.

[0069] Figure 2 It is a module diagram of the ball valve body rotation stop system for a two-way formation isolation valve of the present invention. Specific embodiments

[0070] The embodiments of the present invention will be specifically illustrated below with reference to the accompanying drawings. The given embodiments are only for illustrative purposes and should not be construed as a limitation to the present invention. The accompanying drawings are only for reference and illustration and do not constitute a limitation to the scope of the invention patent protection. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention. In the description of the present invention, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first", "second", "third", etc. may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, the meaning of "a plurality" is two or more.

[0071] In the description of the present invention, it should be noted that unless otherwise clearly specified and limited, the terms "installation", "connection", and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components. The terms "vertical", "horizontal", "left", "right", "up", "down" and similar expressions used herein are only for illustrative purposes and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be construed as a limitation to the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0072] In the description of the present invention, it should be noted that unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used in the description of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0073] As Figure 1 shown, a method for rotating and stopping a spherical valve body of a two-way formation isolation valve provided by an embodiment of the present invention includes:

[0074] S1. Obtain the environmental data of the spherical valve body and the sensors around it, and preprocess all the environmental data to generate an environmental state data set;

[0075] S2. According to the positioning requirements of the spherical valve body, construct a reinforcement learning model, use the environmental state data set as input, and output the optimal positioning parameters of the spherical valve body; wherein, the optimal positioning parameters include: the optimal rotation angle and the optimal positioning torque;

[0076] S3. Extract the current downhole working condition characteristics of the environmental state data set, and adaptively adjust the optimal positioning parameters based on the current downhole working condition characteristics to generate the optimal rotation and stopping torque;

[0077] S4. According to the optimal rotation and stopping torque and the key variables for seal monitoring in the environmental state data set, construct a seal state prediction model, evaluate the seal quality and predict the leakage trend, detect abnormal conditions and perform intelligent fine-tuning to generate the optimal rotation and stopping parameters;

[0078] S5. Calculate the fine-tuning rotation and stopping torque adjustment amount based on the adaptive adjustment target and the evaluation result of the seal quality, and generate the optimal positioning parameters corresponding to the adjustment target based on the optimal rotation and stopping torque and the rotation and stopping torque adjustment amount;

[0079] In a preferred embodiment of the present invention, pressure, vibration, shear force, and torque sensors are arranged around the spherical valve body, and the sensing data is transmitted to the downhole control unit through an optical fiber transmission system. These data are defined as the environmental state data set , where:

[0080] represents the downhole real-time pressure (unit: megapascal).

[0081] represents the vibration amplitude of the spherical valve body (unit: millimeter).

[0082] represents the current torque of the spherical valve body (unit: newton-meter).

[0083] Represents the shear force of downhole fluid (unit: Pa).

[0084] In the embodiments of the present application, data preprocessing: Due to the complex downhole environment, the original sensing data may contain noise and outliers, so data cleaning is required:

[0085] An adaptive sliding window filtering method is used to smooth the continuous data points, and the window size is dynamically adjusted through historical data to ensure signal smoothness:

[0086]

[0087] where, is the filtered data, is the original data at the th time step.

[0088] Set the outlier detection threshold. If , it is determined as an outlier, is a threshold set empirically, and the data points outside the range will be interpolated and compensated.

[0089] Feature extraction and data compression: Due to the limited downhole transmission bandwidth, not all sensing data can be uploaded, so key features need to be extracted:

[0090] Calculate the maximum value, minimum value, mean value and variance within the current data window, and only transmit these statistics to reduce the data volume:

[0091]

[0092]

[0093]

[0094] Only when the data change exceeds the set threshold, the complete time series data is transmitted to reduce the communication burden.

[0095] Data format standardization: To ensure that the data can be directly used by the reinforcement learning model , all data is normalized, and the transformation range is to :

[0096]

[0097] where, is obtained through historical data statistics, is the normalized data.

[0098] Data output: Finally, the preprocessed data is used as the environmental state dataset , input it into the intelligent position finding model , for the intelligent position finding decision of the ball valve body in the next step.

[0099] In the preferred embodiment of the present invention, in S2, a reinforcement learning model is constructed:

[0100] Input the environmental state data and initialize the intelligent position finding model:

[0101] Read the output of S1, that is, the environmental state data set , for intelligent position finding decision, including the following key parameters:

[0102] : downhole pressure (MPa), which affects the force condition of the ball valve body.

[0103] : vibration amplitude of the ball valve body (mm), used to judge the stability of the position finding process.

[0104] : current torque of the ball valve body (N·m), which determines the rotation force of the ball valve body.

[0105] : fluid shear force (Pa), which affects the force state of the ball valve body in the flowing well fluid.

[0106] Due to the complex downhole working conditions, it is difficult for ordinary reinforcement learning models to adapt to the violently changing environment. Therefore, this solution constructs a reinforcement learning position finding model based on dynamic environment perception , and uses to adjust the strategy in real time, so that the ball valve body can accurately dock with the sealing surface under changing downhole conditions.

[0107] Set the rotation angle and the position finding torque of the ball valve body as control variables, and initialize the reinforcement learning intelligent position finding strategy , where is the position finding action.

[0108] Execute the intelligent position finding process: at time step , calculate the current environmental state by , and select the optimal position finding action based on the current strategy :

[0109]

[0110] where is the action value function of the reinforcement learning model.

[0111] After applying the positioning action, collect the docking deviation of the ball valve sealing surface and the sealing pressure , and judge the positioning effect:

[0112] , which is used to evaluate the angular error of the current positioning;

[0113] If (preset tolerance), then adjust and update the strategy;

[0114] If , it means that the positioning torque may be insufficient, increase for compensation.

[0115] Reinforcement learning strategy optimization: Use the state data after positioning to calculate the reward value , and introduce a positioning stability regularization term to optimize the strategy:

[0116]

[0117] Among them, is the docking deviation penalty coefficient (unit: 1 / degree), ensuring that the ball valve body can accurately dock with the sealing surface. is the sealing pressure reward coefficient (unit: MPa), ensuring that the ball valve body can effectively fit the sealing surface after positioning. is the vibration stability penalty term (unit: mm), reducing the oscillation during positioning and improving stability.

[0118] Update the strategy through reinforcement learning, enabling the ball valve body to reach the optimal positioning state faster in future positioning processes, improving stability, and avoiding unnecessary torque adjustments.

[0119] Special adjustment strategies to adapt to complex downhole environments: For high-pressure and high-flow rate working conditions ( , ):

[0120] Enhance the positioning torque compensation to prevent positioning failure caused by fluid impact:

[0121]

[0122] Among them is the torque compensation coefficient, unit: N·m / Pa.

[0123] For low-sealing working conditions downhole ( ):

[0124] Adopt a step-by-step positioning method and gradually adjust until the sealing pressure reaches the target value:

[0125]

[0126] wherein is the positioning step coefficient, unit: degree / MPa.

[0127] Output the optimal positioning parameters:

[0128] When and the positioning is successful, output the optimal positioning parameters:

[0129] Optimal rotation angle

[0130] Optimal positioning torque

[0131] These parameters will be input into the adaptive rotation stop torque control in step S3 to ensure that the rotation stop process starts from accurate positioning and improve the sealing reliability.

[0132] In a preferred embodiment of the present invention, S3 includes:

[0133] During the rotation stop process of the ball valve body, the accuracy of torque control directly determines the sealing reliability. Traditional methods usually use a fixed torque value, but this method cannot adapt to complex downhole environments and may cause damage to the sealing surface or poor sealing. This step is based on the adaptive rotation stop torque control model , combined with real-time downhole working condition data, to intelligently adjust the torque. Through environmental state perception, dynamic torque calculation and real-time feedback optimization, it is ensured that the ball valve body achieves the best sealing effect during the rotation stop process and avoids sealing failure caused by pressure mutation or fluid interference.

[0134] Among them, input the optimal positioning parameters and initialize the adaptive rotation stop torque model:

[0135] Read the output of step 2, that is, the optimal positioning parameters:

[0136] : Optimal rotation angle (degree) to ensure that the ball valve body rotates and stops at the correct position; : Optimal positioning torque (N·m) to provide an initial rotation stop torque reference.

[0137] Combine the environmental state data set in step S1 , and extract the current downhole working conditions:

[0138] : Downhole pressure (MPa), which affects the stability of the sealing state.

[0139] : Vibration amplitude of the ball valve body (mm), used to judge the dynamic stability during the rotation and stopping process.

[0140] : Fluid shear force (Pa), which determines the degree of interference of the fluid on the rotation and stopping process.

[0141] Furthermore, during the traditional rotation and stopping process of the ball valve body, the torque often adopts a fixed value, ignoring the dynamic changes of downhole conditions, which may lead to damage or poor sealing of the sealing surface. This solution proposes an adaptive rotation and stopping torque control model based on multi-factor dynamic adjustment , which can optimize the torque adjustment strategy according to real-time feedback and improve the sealing reliability.

[0142] Furthermore, calculate the initial rotation and stopping torque and apply control: set the initial torque for the ball valve to rotate and stop , and perform dynamic correction based on the environmental state data to ensure that the initial torque adapts to the current downhole conditions:

[0143]

[0144] Wherein, is the standard downhole pressure (MPa), is the reference fluid shear force (Pa). and are the pressure and shear force compensation coefficients respectively (N·m / MPa, N·m / Pa).

[0145] Furthermore, apply to rotate and stop the ball valve body, and monitor the sealing state parameters in real time: : Sealing pressure (MPa), used to judge the sealing state.

[0146] : Leakage detection signal (0 or 1), used to judge whether there is seal failure.

[0147] Furthermore, adopt a dynamic oscillation suppression strategy to prevent damage to the sealing surface caused by fluid disturbance by adjusting the torque application rate.

[0148] Adaptive adjustment of the rotation and stopping torque: Monitor the sealing state parameters. If is lower than the set threshold or (there is leakage), then adjust the torque:

[0149]

[0150] Wherein, is the sealing pressure compensation coefficient (N·m / MPa). is the leakage correction coefficient (N·m). Calculate the final optimal rotation and stopping torque:

[0151]

[0152] Apply the upper limit constraint of torque to avoid damage to the sealing surface due to excessive pressure.

[0153] Output the optimal screwing torque and transmit it to the seal monitoring:

[0154] When it reaches the set value and there is no leakage signal, the screwing is successful, and the output is:

[0155] Optimal screwing torque .

[0156] This torque will be input into the seal state monitoring in step S4 to judge the sealing quality after screwing and ensure long-term sealing stability.

[0157] In the preferred embodiment of the present invention, S4 includes:

[0158] After the ball valve body is screwed, the stability of the seal state directly determines the isolation effect. Due to the complex downhole environment, the sealing effect may be affected by factors such as downhole pressure changes, fluid erosion, and material micro-deformation, resulting in a decrease in sealing performance or leakage. This step proposes a seal state monitoring and intelligent evaluation model , which is used to monitor the seal pressure, vibration, and leakage signal in real time, and optimize the seal state by combining historical data and intelligent analysis. This solution ensures that the ball valve body can maintain long-term stable isolation under different downhole working conditions through dynamic pressure adjustment, seal state prediction, and abnormal identification.

[0159] Input the optimal screwing torque and initialize the seal state monitoring model:

[0160] Read the output of step 3, that is, the optimal screwing torque:

[0161] : The torque (N·m) finally used to screw the ball valve body, which affects the pressing degree of the sealing surface.

[0162] Combine the environmental state data set in step 1 , and extract the key variables for seal monitoring:

[0163] : Seal pressure (MPa), which is used to judge the stability of the seal state.

[0164] : Leakage detection signal (0 or 1), which is used to judge whether there is leakage on the sealing surface.

[0165] : Vibration amplitude of the ball valve body (mm), which is used to evaluate the stability of the sealing surface.

[0166] Due to the dynamic changes in the influence of downhole pressure, temperature, and fluid shear force on the sealing state, this solution adopts an adaptive pressure adjustment mechanism to ensure the adaptability of the sealing threshold under different downhole conditions.

[0167] Construct a sealing state monitoring and intelligent evaluation model: Set the target pressure range of the sealing state, and dynamically adjust the sealing pressure threshold based on historical data:

[0168]

[0169] Among them, is the standard sealing pressure (MPa), set based on laboratory test data. is the downhole pressure influence coefficient (dimensionless), used to adjust the sealing judgment standard according to the environment.

[0170] Furthermore, adopt a sealing state prediction model to comprehensively , , Calculate the sealing quality score , and introduce a leakage risk regularization term to quantify the sealing state:

[0171]

[0172] Among them, are the weights of the sealing pressure, leakage signal, vibration amplitude, and leakage risk factor (unit: MPa, binary, mm, dimensionless), respectively. is the predicted value of the leakage trend, calculated based on the time series change rate:

[0173]

[0174] Among them is the pressure deviation amplification coefficient (unit: 1 / MPa), ensuring early warning of leakage risk near the critical point of the sealing state.

[0175] Furthermore, set the sealing state judgment standard:

[0176] If , the sealing state is good, and enter the end state.

[0177] If , there are potential hazards in the sealing state, and enter the intelligent optimization stage.

[0178] If , the seal fails and the rotation stop torque needs to be readjusted.

[0179] Furthermore, intelligent anomaly detection and sealing optimization suggestions:

[0180] If If the leakage risk exceeds the set value, an intelligent early warning is triggered, and the changing trend of the sealing parameters is analyzed:

[0181] Calculate to determine whether the sealing performance is gradually decreasing.

[0182] Furthermore, an abnormal sealing state classification model is adopted to judge the type of seal failure based on the long-term trend and short-term changes:

[0183] If<00.org / 10.1007 / s11227-022-04201-4" target="_blank">https: / / doi.org / 10.1007 / s11227-022-04201-4 gradually decreases, and increases, it is possible that material fatigue occurs on the sealing surface, and the rotation stop torque needs to be increased.

[0184] If suddenly decreases, it is possible that microcracks are generated on the sealing surface, and intelligent fine-tuning needs to be performed.

[0185] Furthermore, according to the determination result of the sealing state, optimization suggestions are given: <org / 10.1007 / s11227-022-04201-4" target="_blank">https: / / doi.org / 10.1007 / s11227-022-04201-4

[0186] If the torque needs to be adjusted, input the intelligent fine-tuning optimization of step S5 and recalculate the optimal rotation stop parameters.

[0187] If the sealing state is stable, output the final sealing state evaluation result.

[0188] Furthermore, output the sealing state evaluation result:

[0189] When and the sealing state is stable, output:

[0190] Sealing state evaluation index for recording the sealing quality.

[0191] If is lower than the set threshold value, input it into the intelligent fine-tuning optimization of step 5 to further adjust the sealing state and ensure long-term stable isolation.

[0192] In the preferred embodiment of the present invention, S5 includes:

[0193] Among them, after the sealing state is monitored, if the sealing state does not meet the set stable standard, intelligent fine-tuning optimization is required to ensure that the ball valve body maintains effective isolation for a long time. In this step, through sealing state feedback, adaptive parameter adjustment and dynamic optimization control, the adjustment amount is calculated in real time according to the sensor data, and the fine-tuning operation is performed. This solution focuses on solving the problems of insufficient sealing, sealing overload and leakage trend, and ensures the adaptability and stability of the sealing performance in complex downhole environments.

[0194] Input the sealing state evaluation result and initialize the fine-tuning optimization model:

[0195] Read the output of step S4, i.e., the sealing state evaluation index:

[0196] : Sealing quality score (dimensionless), used to quantify the sealing effect.

[0197] Combine the environmental state data set in step 1 , and extract the key variables for seal adjustment:

[0198] : Sealing pressure (MPa), used to judge the sealing state.

[0199] : Leak detection signal (0 or 1), used to judge whether there is leakage on the sealing surface.

[0200] : Optimal stop torque (N·m), used to adjust the sealing fit state.

[0201] If , it means that the sealing state needs to be optimized, and start the seal fine-tuning optimization model for intelligent adjustment.

[0202] Furthermore, calculate the fine-tuning adjustment amount:

[0203] Set the adaptive adjustment target:

[0204] If , it means that the sealing pressure is insufficient and the stop torque needs to be increased .

[0205] If and is too large, there may be overload on the sealing surface or uneven mechanical stress, and it is necessary to reduce and adjust the rotation angle .

[0206] Furthermore, calculate the fine-tuning stop torque adjustment amount:

[0207]

[0208] Among them, is the sealing pressure correction coefficient (N·m / MPa). is the vibration influence correction coefficient (N·m / mm).

[0209] Calculate the fine-tuning rotation angle amount:

[0210]

[0211] Among them, is the leakage signal trigger adjustment coefficient (degree). is the leakage trend correction coefficient (degrees / second). is the predicted value of the leakage trend (dimensionless), calculated based on time series analysis.

[0212] Furthermore, perform fine-tuning optimization:

[0213] Apply the adjusted rotation stop torque:

[0214]

[0215] Apply the adjusted rotation angle:

[0216]

[0217] After applying the fine-tuning, re-monitor the sealing state and determine whether it meets . If the target is still not achieved, iteratively execute the fine-tuning process until the seal is stable.

[0218] Finally, output the optimized sealing state parameters:

[0219] If , output:

[0220] Optimized rotation stop torque .

[0221] Optimized rotation angle .

[0222] Record the final sealing state to ensure long-term stable isolation, and store it in the downhole data storage unit for future analysis and optimization.

[0223] Correspondingly, as Figure 2 shown, based on a ball valve body rotation stop method of a two-way formation isolation valve, an embodiment of the present invention further provides a ball valve body rotation stop system for a two-way formation isolation valve, including:

[0224] A ball valve body data acquisition unit 1, configured to acquire environmental data of the ball valve body and its surrounding sensors, and preprocess all environmental data to generate an environmental state data set;

[0225] A positioning parameter analysis unit 2, configured to construct a reinforcement learning model according to the positioning requirements of the ball valve body, take the environmental state data set as input, and output the optimal positioning parameters of the ball valve body; wherein, the optimal positioning parameters include: optimal rotation angle and optimal positioning torque;

[0226] A rotation stop torque analysis unit 3, configured to extract the current downhole working condition characteristics of the environmental state data set, and adaptively adjust the optimal positioning parameters based on the current downhole working condition characteristics to generate the optimal rotation stop torque;

[0227] The intelligent fine-tuning unit 4 is used to construct a seal state prediction model according to the optimal stop torque and the key variables for seal monitoring in the environmental state dataset, evaluate the seal quality and predict the leakage trend, detect abnormal conditions and perform intelligent fine-tuning to generate optimal stop parameters;

[0228] The ball valve body stop unit 5 is used to calculate the fine-tuning stop torque adjustment amount based on the adaptive adjustment target and the evaluation result of the seal quality, and generate the optimal positioning parameters corresponding to the adjustment target based on the optimal stop torque and the stop torque adjustment amount;

[0229] Wherein, the adaptive adjustment target is seal pressure adjustment.

[0230] Unless otherwise specifically stated, the relative steps, numerical expressions, and values of the components and steps set forth in these embodiments do not limit the scope of the present application.

[0231] If the above functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium executable by a processor. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the system described in each embodiment of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0232] In the description of the present application, it should be noted that the orientation or positional relationship indicated by terms such as "upper" and "lower" is based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the inventive product is usually placed during use. It is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present application.

[0233] In the description of the present application, it should also be noted that unless otherwise clearly defined and limited, the terms "set", "installed", "connected", and "coupled" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0234] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for rotating and stopping a ball valve body of a two-way formation isolation valve, characterized in that, The method includes: Obtaining the environmental data of the ball valve body and the surrounding sensors, and preprocessing all the environmental data to generate an environmental state data set; Constructing a reinforcement learning model according to the positioning requirements of the ball valve body, taking the environmental state data set as the input, and outputting the optimal positioning parameters of the ball valve body; wherein, the optimal positioning parameters include: the optimal rotation angle and the optimal positioning torque; Extracting the current downhole working condition characteristics of the environmental state data set, and adaptively adjusting the optimal positioning parameters based on the current downhole working condition characteristics to generate the optimal rotation stop torque; Constructing a seal state prediction model according to the optimal rotation stop torque and the key variables for seal monitoring in the environmental state data set, evaluating the seal quality and predicting the leakage trend, detecting abnormal conditions and performing intelligent fine-tuning to generate the optimal rotation stop parameters; Calculating the fine-tuning rotation stop torque adjustment amount based on the adaptive adjustment target and the evaluation result of the seal quality, and generating the optimal positioning parameters corresponding to the adjustment target based on the optimal rotation stop torque and the rotation stop torque adjustment amount; Wherein, the adaptive adjustment target is seal pressure adjustment.

2. The ball valve body rotation stop method of a two-way formation isolation valve according to claim 1, characterized in that, The environmental data includes the downhole real-time pressure, the vibration amplitude of the ball valve body, the current torque of the ball valve body, and the shear force of the downhole fluid; the preprocessing includes: Obtaining the original environmental data; Smoothing the continuous data points of the original environmental data, wherein the window size is dynamically adjusted by historical data; Detecting the outliers in the data after smoothing processing, and interpolating and compensating the data points beyond the range to obtain the corrected data; Extracting the key features of the corrected data and performing normalization processing; wherein, the key features include the maximum value, minimum value, mean value, and variance within the current data window; Obtaining the preprocessed data and constructing it into an environmental state data set.

3. The ball valve body rotation stop method of a two-way formation isolation valve according to claim 2, characterized in that, The maximum value, minimum value, mean value, and variance within the current data window include: The maximum value of the downhole real-time pressure, the minimum value of the downhole real-time pressure, the mean value of the downhole real-time pressure, and the variance of the downhole real-time pressure of the downhole real-time pressure; the maximum value of the vibration amplitude of the ball valve body, the minimum value of the vibration amplitude of the ball valve body, the mean value of the vibration amplitude of the ball valve body, and the variance of the vibration amplitude of the ball valve body, the current torque of the ball valve body; the maximum value of the current torque of the ball valve body, the minimum value of the current torque of the ball valve body, the mean value of the current torque of the ball valve body, and the variance of the current torque of the ball valve body.

4. The ball valve body rotation stopping method of a two-way formation isolation valve according to claim 2, characterized in that, The constructing of the reinforcement learning model according to the positioning requirements of the ball valve body specifically includes: Obtaining the environmental state data set, constructing a reinforcement learning model, and adjusting the strategy when using the environmental state data set so that the ball valve body can accurately dock with the sealing surface under changing downhole conditions; setting the rotation angle and positioning torque of the ball valve body as control variables, and initializing the reinforcement learning intelligent positioning strategy of the reinforcement learning model; At the current time step, calculating the current environmental state by the reinforcement learning model, and selecting the optimal positioning action based on the current strategy. After applying the positioning action, collecting the docking deviation of the ball valve seal surface and the seal fitting pressure, and judging the positioning effect; Calculating the reward value based on the state data after positioning, and introducing the positioning stability regularization term to optimize the positioning strategy; Wherein, the positioning strategy satisfies the following constraints: A. For high-pressure and high-flow-rate working conditions, it is necessary to enhance the positioning torque compensation to prevent positioning failure caused by fluid impact. B. For low-sealing conditions underground, a step-by-step positioning method needs to be adopted to gradually adjust the rotation angle until the sealing pressure reaches the target value. Among them, when the docking deviation of the ball valve sealing surface is less than or equal to the preset tolerance and the sealing fitting pressure is greater than or equal to the sealing fitting pressure threshold value, the positioning is successful, and the optimal positioning parameters are output: the optimal rotation angle and the optimal positioning torque.

5. A method for rotating and stopping the ball valve body of a two-way formation isolation valve according to claim 4, characterized in that, Collect the docking deviation and sealing fitting pressure of the ball valve sealing surface to judge the positioning effect. Specifically: If the docking deviation of the ball valve sealing surface is equal to the difference between the current rotation angle and the target rotation angle, it is directly used to evaluate the angle error of the current positioning. If the docking deviation of the ball valve sealing surface is greater than the preset tolerance, adjust the rotation angle and update the strategy. If the sealing fitting pressure is less than the sealing fitting pressure threshold value, it indicates that the positioning torque may be insufficient, and the positioning torque needs to be increased for compensation.

6. A method for rotating and stopping the spherical valve body of a two-way formation isolation valve according to claim 4, characterized in that, Adaptive adjustment of the optimal positioning parameters based on the current underground working condition characteristics to generate the optimal rotation-stop torque, specifically including: Set the initial torque for the ball valve to stop , and perform dynamic correction based on environmental state data to ensure that the initial torque is suitable for the current downhole working conditions, and the calculation is as follows: Among them, is the standard downhole pressure, is the reference fluid shear force; and are the pressure and shear force compensation coefficients respectively; Apply torque to the rotary stop ball valve body and monitor the seal state parameters in real time: is the shear force of the downhole fluid; Sealing pressure , used to determine the sealing state; Leak detection signal , used to determine whether there is a seal failure; Monitor the seal status parameters. If the seal pressure is lower than the set threshold or the leak detection signal , indicating a leak, then adjust the torque: wherein, is the torque adjustment amount, is the seal pressure compensation coefficient; is the leakage correction coefficient; is the seal fitting pressure threshold value; is the seal fitting pressure; Calculate the final optimal rotation-stop torque by combining the initial torque and the torque adjustment amount. When the sealing pressure reaches the set value and there is no leakage signal, the rotation-stop is successful, and the optimal rotation-stop torque is output.

7. A method for rotating and stopping the ball valve body of a two-way formation isolation valve according to claim 6, characterized in that, The sealing state prediction model is based on the leakage risk regularization term for prediction to ensure early warning of leakage risk near the sealing state critical point and output the sealing quality score. If the sealing quality score is greater than the set value, trigger an intelligent warning and analyze the change trend of the sealing parameters: Calculate the change amount of the sealing pressure and judge whether the sealing performance is gradually decreasing. Judge the sealing failure type based on the long-term trend and short-term changes. If the change amount of the sealing pressure gradually decreases and the leakage detection signal increases, the sealing surface may undergo material fatigue, and the rotation-stop torque needs to be increased. If the change amount of the sealing pressure suddenly decreases and the leakage detection signal is 1, the sealing surface may generate microcracks, and intelligent fine-tuning needs to be performed to recalculate the optimal rotation-stop parameters.

8. A method for rotating and stopping the ball valve body of a two-way formation isolation valve according to claim 7, characterized in that, The intelligent fine-tuning and recalculation of the optimal rotation-stop parameters specifically include: Set the adaptive adjustment target: if the sealing fitting pressure is less than the sealing fitting pressure threshold value, the rotation-stop torque needs to be increased; if the sealing fitting pressure is greater than the sealing fitting pressure threshold value and the leakage detection signal is too large, there may be overload of the sealing surface or uneven mechanical stress, and the rotation-stop torque needs to be reduced and the rotation angle needs to be adjusted. Calculate the fine-tuning rotation stop torque adjustment amount : Among them, is the sealing pressure correction coefficient; is the vibration influence correction coefficient; is the sealing fit pressure threshold value; is the sealing fit pressure; is the leak detection signal; calculate the fine adjustment amount of the rotation angle : Among them, is the leakage signal trigger adjustment coefficient; is the leakage trend correction coefficient; is the predicted value of the leakage trend, calculated based on time series analysis; Apply the adjusted rotation-stop torque and the adjusted rotation angle, re-monitor the sealing state, and judge whether the target is reached, then iteratively execute the fine-tuning process until the sealing is stable.

9. A ball valve body rotation stop system for a two-way formation isolation valve, characterized in that, The system includes: A ball valve body data acquisition unit for acquiring the environmental data of the ball valve body and its surrounding sensors, and preprocessing all the environmental data to generate an environmental state data set. A positioning parameter analysis unit for constructing a reinforcement learning model according to the positioning requirements of the ball valve body, taking the environmental state data set as input, and outputting the optimal positioning parameters of the ball valve body; among them, the optimal positioning parameters include: the optimal rotation angle and the optimal positioning torque. A rotation stop torque analysis unit, which is used to extract the current downhole working condition characteristics of the environmental state data set, adaptively adjust the optimal positioning parameters based on the current downhole working condition characteristics, and generate the optimal rotation stop torque; An intelligent fine-tuning unit, which is used to construct a seal state prediction model according to the optimal rotation stop torque and the key variables for seal monitoring in the environmental state data set, evaluate the seal quality and predict the leakage trend, detect abnormal conditions and perform intelligent fine-tuning to generate the optimal rotation stop parameters; A ball valve body rotation stop unit, which is used to calculate the fine-tuning rotation stop torque adjustment amount based on the adaptive adjustment target and the evaluation result of the seal quality, and generate the optimal positioning parameters corresponding to the adjustment target based on the optimal rotation stop torque and the rotation stop torque adjustment amount; Wherein, the adaptive adjustment target is seal pressure adjustment.

Images